2021:Unconference/Wikimedia Research Community Session

Wikimedia Research is a team of scientists and engineers using data to understand and empower millions of readers and contributors who interact with Wikipedia and its sister projects on a daily basis.

We are running an unconference session to connect with community members who are interested in learning more about our work and / or conducting their own research on Wikimedia projects. Session notes (etherpad)

We have three research programs:

• Address knowledge gaps:
  • We are building a Knowledge Gaps Index, which can help understand the extent of knowledge gaps across Wikimedia projects through quantifiable evidence.
  • We work on building algorithms for Structured Tasks , such as the Link Recommendation Algorithm, and the Image Suggestion Algorithm
  • We prototype metrics to define article importance, such as Misalignment (distance between article quality and article popularity)
  • We work on understanding readers' motivations, navigation patterns, and image usage.
• Improve knowledge integrity
  • We work on understanding disinformation on Wikipedia
  • We built the Knowledge Integrity Risk Observatory to monitor indicators of threats to Wikimedia projects.
  • We build datasets for researchers to investigate Reliability on Wikipedia
• Conduct foundational work
  • We built language agnostics classifiers for article quality, article topic, and article categorization by country
  • We organize events for the Wikimedia Research community such as Wiki Workshop, and data challenges such as the TREC Fair ranking task

We are generally interested in (random order):

• Metrics and indicators to support Knowledge Equity and Knowledge Integrity
• Machine learning technologies for automated content tagging (images and text)
• Fair and inclusive Recommender Systems
• Structured Data Challenges
• Growing a community of Wikimedia researchers
• Open research datasets
• Understanding controversial content
• Challenges of fact checking using Wikimedia data
• Wikimedia data reuse outside our sites
• How readers use, navigate and learn through Wikipedia
• Large-scale data analysis to understand and serve Wikimedia communities

Contact

LZia (WMF) or ELescak (WMF)

Sign-up here:

Please add your wikipedia username below! You're also welcome to add any questions / topics you would like to discuss. We look forward to seeing you!